Chapter 2: Public Perceptions of Crime

This section presents findings from the 2013/14 Crime Survey for England and Wales (CSEW) looking at people’s perception of crime in their local area and in the country as a whole. It also examines people’s perceived likelihood of victimisation and their worry about crimes. Unless otherwise stated, all changes over time and differences in measures described were found to be statistically significant¹ at the 5% level.

• Around 6 in 10 adults (61%) perceived crime in the country as a whole to have risen over the past few years. Fewer people, however, perceived that crime had risen in their local area (32%)

• When asked about the level of crime in their local area, compared with the level nationally, only a small proportion (9%) thought crime in their local area was above average and 55% felt it was below average

• News programmes on TV and radio were most often cited as a source of information that had influenced perceptions of national crime levels (cited by 67% of people). In contrast, word of mouth was most often cited as a source influencing impressions of local crime rates (cited by 54%)

• There were 12% of adults classified as having a high level of worry about violent crime, 11% about burglary and 7% about car crime. All of these measures were at a similar level to the previous year and the general trend has been flat for a number of years

• There were 19% of adults who thought it was either “very” or “fairly likely” that they would be a victim of crime within the next 12 months

• Perception of local crime had a stronger relationship with perceived likelihood of victimisation, than perception of national crime levels. For those who felt crime had been rising locally, 33% thought they were either “very” or “fairly likely” to be a victim in the next year. This is compared with 23% who felt crime had risen in the country as a whole.

• In line with previous surveys, those who had been a victim of crime in the last year were more likely to: consider both local and national crime rates to have risen over the last few years; have a high level of worry about crime; and to think it likely they would suffer victimisation in the year ahead, than non-victims

Notes: Section 8 of the User Guide provides more information on how this is calculated

As well as being used to estimate levels of crime in England and Wales, the Crime Survey for England and Wales (CSEW) also contains a range of questions on respondents’ perceptions of crime. These include questions on perceived levels of crime in the country as a whole and in the local area, worry about crime and perceived likelihood of victimisation. This chapter presents analyses of these questions based on responses to the 2013/14 CSEW looking at variations by household, personal and area characteristics.

While the level of crime measured by the CSEW has been falling since a peak in 1995, the survey has consistently shown that most people perceive that crime across the country as a whole has still been rising. This contrast has continued with the 2013/14 survey showing 61% of adults thinking crime had gone up nationally in the last few years. However, the proportion of adults holding this belief is the lowest it has been since 2008/09 survey when this question was introduced into the survey ( Appendix table 2.01 (733.5 Kb Excel sheet) )¹.

While 6 in 10 adults thought crime had been rising across the country as whole, only 1 in 3 (32%) thought it had done so in their local area ( Appendix Table 2.01 (733.5 Kb Excel sheet) ). In addition, when asked about the comparative level of crime in their local area, only a small proportion of adults (9%) thought that crime in their local area was “above average”, with many more (55%) thinking that it was “below average” (Table 2.4).

Figure 2.1 shows the relationship between people’s perceptions of the comparative level of crime in their local area and the Crime Domain of the English Indices of Deprivation (2010). The latter is based on the level of police recorded crime per head of population and is an objective measure of crime². The chart indicates a clear relationship between objective levels of crime in a local area³ and subjective perceptions. For example, the proportion of adults perceiving their area to have lower than average levels of crime rises as recorded crime levels fall. Similarly, those living in the higher crime areas were more likely to give the response that the local crime rate was “about average” or “higher than average” (Figure 2.1).

Previous sweeps of the survey have also highlighted a gap between perceptions of crime levels locally and in the country as a whole. Duffy et al., (2008) point to data collated on views of the NHS, as well as to data relating to views about crime levels in North America, Australia and South Africa, and note that this gap between local and national perceptions is neither new nor unique to crime, or to Britain. Duffy et al., go on to suggest a number of possible explanations for the gap between perceptions of local and national crime levels. These include: the potentially greater influence the media may have views of national as compared to local crime; and the possibility that people show a natural ‘hometown favouritism’ to their local area.

Notes for section 1 - Perceptions of changing local and national crime levels

1. From 2008/09 the survey has included the question ‘What do you think has happened to crime in the country as a whole over the past few years?’ Until 2011/12 this was asked in addition to the question it eventually replaced: ‘How much would you say the crime rate here has changed since two years ago? In this area, would you say there is more crime or less crime?’ Possible responses to both questions are on a 5-point scale ranging from gone up a lot, to gone down a lot.

2. There is more information in Section 7.1 of the User Guide to Crime (1.61 Mb Pdf) Statistics for England and Wales on how the crime domain of the English Indices of Deprivation (2010) is calculated. Wales is excluded.

3. Local area definitions for the English Indices of deprivation are based on Super Output Areas (SOAs). These are a set of geographical areas developed following the census to produce a set of areas of consistent size. Lower Layer Super Output Areas (LSOAs) typically have a population of around 1500.

This next section examines how perceptions of crime levels varied by a range of personal, household and area characteristics.

Women were more likely than men to have believed crime had risen in recent years. This was true for both crime across the country as a whole (68% and 55%, respectively) and locally (36% and 28% respectively).

There were also differences by age group in perceptions of crime levels but the patterns were different for the local and national measures. With regard to the national picture, the general pattern was for perceptions of crime rising to rise with age; those aged 75 and over were most likely to view it as having risen in recent years (72%). Conversely perceptions of crime rising locally peaked in the 35 to 44 and 45 to 54 age groups (36% and 37% respectively), and then reduced with rising age. While those aged 75 and over were the most likely to think crime was rising across the country as whole, they were the least likely to view local crime as having risen (25%), (Figure 2.2).

Personal and household characteristics associated with relative disadvantage were linked to increased likelihood of perceiving crime to be rising. This was generally true for perceptions of changes to both local and national levels of crime¹.

• Those with an annual household income of less than £10,000 were more likely to believe crime had risen in the country as a whole (71%) and in their local area (37%), than other groups. For example, for those in the highest household income group (£50,000 or more) only 47% were more likely to believe crime had risen in the country as a whole, and 29% in their local area ( Appendix Table 2.03 (733.5 Kb Excel sheet) ).

• Social renters were more likely than people in other tenures to believe both crime in the country as a whole had been rising (71% compared with 61% of home owners and 56% of private renters) and locally (36% compared with 31% of home owners and 32% of private renters) ( Appendix Table 2.03 (733.5 Kb Excel sheet) ).

• Individuals living in the 20% most deprived areas (as measured by the employment domain of the English Indices of Deprivation) were most likely perceive a rise in crime over the last few years both nationally and locally (68% and 39% respectively, compared with 55% and 26% for the least deprived output areas) ( Appendix Table 2.03 (733.5 Kb Excel sheet) ).

Conversely, the following factors were all much more strongly related to views of changing levels of local crime, but less so for national crime:

• adults who lived in an urban area were more likely to think that crime had risen in their local area than those resident in a rural area (34% compared with 26%)

• those living in areas of high physical disorder² were more likely than those living in areas of low physical disorder, to believe local crime had risen (41% compared with 32%)

Figure 2.3 presents a more detailed analysis of people’s views of crime in the context of their local area by using the observational data on the level of physical disorder in the immediate area around respondents’ homes gathered by survey interviewers. This generally shows that, controlling for the level of crime in the area (as indicated by the crime domain of the English Indices of Deprivation), the higher the level of assessed physical disorder in an area, the more likely it was for people to say crime in their local area was “higher than” or “about the same” as crime in the country as a whole. This was true across all the deciles except in the 10% least deprived LSOA’s (Figure 2.3).

Notes for perceptions of changing crime levels by background characteristics

1. Of course, a number of these measures will be inter-related and also associated with the actual level of crime in an area.

2. Physical disorder is a measure based on the interviewer’s assessment of the level of (a) vandalism, graffiti and deliberate damage to property; (b) rubbish and litter; and (c) homes in poor condition in the area (Section 7.1, User Guide).

Following questions on their perceptions of crime over the past few years, respondents were asked what had informed their perception. In line with previous findings, the 2013/14 CSEW showed that adults most often cited news programmes on television or radio as the source that informed their impression of levels of crime in the country as a whole (67%). Other common sources mentioned were word of mouth or information from other people (32%), and newspapers (‘tabloid’ 31% and ‘local’ 31%), (Table 2.1).

Table 2.1 shows that around a third of adults said that “tabloid” newspapers and, 1 in 4 that “broadsheet” newspapers, were a source that informed their perception of what was happening to crime across the country as a whole.

CSEW respondents were also asked questions about their use of newspapers¹, including which newspaper they had read most often within the last 3 months. If an individual had read more than 1 newspaper they were asked which one they preferred. As seen in table 2.3, the 2013/14 CSEW showed that 70% of adults who preferred to read a “popular” newspaper thought crime in the country had risen in last few years, while only 10% thought it had fallen. A lower percentage of “broadsheet” readers believed crime had risen (43%) and twice as many thought it had gone down (21%).

It is likely that part of the explanation for the better fit between subjective perceptions and objective levels of crime in the local area is because people are able to make judgements based on their own experience of their local area (Table 2.2). In contrast, when making judgements about the level of crime in the whole country, people rely much more heavily on the media. It has been suggested that reliance on the media for information, is linked to a misconception that crime is rising (Pfeiffer et al., 2005). Additionally, Duffy et al., (2008) found that when the public were asked why they think there is more crime than 2 years ago, almost half said it was because of what they read in the papers. Newspapers readership was predictive of feelings about crime in a regression analysis. This was the case, even after controlling for voting patterns and the authors suggest their analysis provides a strong case for the direct impact of newspaper content on views (Duffy et al., 2008).

Notes for sources of information on crime

1. These questions will also be asked in the 2014/15 survey but ONS plan to remove them from the 2015/16 survey to make space for new questions.

In line with previous surveys, perceptions of crime rates were also associated with personal experiences of crime. Previous analysis has shown that victims of crime were more likely to think that crime had increased locally over the last few years than those who were not victims (Flatley et al., 2010).

This analysis is supported by the 2013/14 CSEW which showed that adults who had been a victim of crime in the last year were more likely to view their local crime rates as having risen over the last few years (46% compared with 30% of non-victims), ( Appendix Table, 2.02 (733.5 Kb Excel sheet) ). There was also a difference in perceptions of the national crime rate, albeit less pronounced (66% of victims believed crime had increased in the country as a whole, compared with 61% of non-victims). However, this general pattern masked the variation in perceptions by the location of victimisation.

Table 2.4 shows adults who had been victims of crime in their local area¹ were more than twice as likely as non-victims to say that the local crime rate was higher than the average for the country (19% compared with 8%). In contrast, for those who had been victims of crime elsewhere in England and Wales, there was no difference in the proportion of victims and non-victims, who said the local crime rate was above average (estimated to be 8% for both groups).

In addition, adults who had been victims of crime in their local area were less likely to say that the local crime rate was below average (40%) compared with those who had been victims elsewhere in England and Wales or non-victims (58% and 57% respectively; Table 2.4).

Notes for experience of victimisation

1. Numbers for victims of crime in the local area are calculated from victims responding positively to a question asking if the incident took place in their resident area (within 15 minutes walk of their home address).

Worry about crime

The 2013/14 CSEW shows that 12% of adults were classified as having a high level of worry about violent crime, 11% about burglary, and 7% about car crime¹. All of these measures were at a similar level to the previous year and the general trend has been flat for a number of years. However, all measures are significantly lower than in the mid-to-late 1990s when crime measured by the CSEW was at much higher levels than today ( Appendix Table 2.06 (733.5 Kb Excel sheet) ).

Figure 2.4 shows the relationship between worry about crime and age. This age distribution was similar to that for perceptions of local crime levels (Figure 2.2). It seems that those aged between 45 and 64 were generally more worried about crime than other age groups. Aside from the notable exception being worry about violent crime, where those in the youngest age group (16 to 24) had the highest level of worry.

It has been claimed that questions about worry tap into feelings of perceived vulnerability factors (Heath et al, 2001). Therefore, variation between groups may reflect more differences in perceived vulnerability rather than the perceived risk of falling victim. Rountree and Land (1996) suggested that responses to questions about worry may encompass more than just worry about the crime being committed and, also include concerns about the consequences of different crime types.

A broader conceptualisation of worry, might partly explain why women had a higher level of worry about being a victim of violent crime than men (18% compared with 6%). The pattern was similar for worry about burglary where 14% of women had a high level of worry compared with only 8% of men. In contrast, there was no difference between the sexes in relation to worry about car crime, where 7% of both men and women had high levels of worry² ( Appendix Table 2.07 (733.5 Kb Excel sheet) ).

Additional variations of interest in relation to worry about crime are listed here.

• Those with a long standing disability were more likely to have a high level of worry about burglary (15%) and violent crime (15%), compared with those without a longstanding disability (10% and 11% respectively).

• Social renters were more likely than those with other forms of tenure to be worried about all crime types (15% were worried about burglary, 12% about car crime and 18% about violent crime), particularly in comparison to levels of worry for owner occupiers (10% were worried about burglary, 6% about car crime and 10% about violent crime) ( Appendix Table 2.08 (733.5 Kb Excel sheet) ).

• Residents in areas of high physical disorder were more likely to be worried about both burglary (17%) and violent crime (21%), than those who were not in an area of high physical disorder (11% and 12% respectively), although this is not true for worry about car crime.

• Residents in the 20% most deprived areas, were more worried about all crime types than individuals in all other areas, particularly for worry about violent crime, where 18% were “very” or “fairly” worried, as compared to only 7% of those in the 20% least deprived areas.

• Those living in urban areas were more likely to be worried about all crime types measured than those living in rural areas; with the greatest difference being recorded for worry about violent crime (13% compared with 7%).

• Victims were more likely to worry than non-victims about; burglary (16% compared with 10%), car crime (11% compared with 6%) and violent crime (15% compared with 11%, respectively).

Perception of changes in both national and local crime levels was generally associated with worry about crime. For example, those who thought local crime had increased over the past few years were much more likely to be worried about burglary (18%), than those who did not (8%). This association was also found in relation to both violent crime (17% and 9%, respectively) and car crime (10% and 5%, respectively) (Table 2.5).

Notes for section 2 - Worry about crime and perceived likelihood of victimisation

1. Section 6.2 of the User Guide (1.59 Mb Pdf), provides further detail on how worry about violent crime, burglary and car crime is measured by the CSEW

2. Questions relating to the perceived likelihood of being a victim of vehicle crime are only asked of vehicle owners.

The CSEW also asks people about “how likely do you think you personally are to be a victim of crime in the next year”. It was believed by 19% of adults that it was either “very” or “fairly likely” that they would be a victim of crime within the next year ( Appendix Table 2.04 (733.5 Kb Excel sheet) ).

Perceived likelihood of victimisation was associated more strongly with perceptions of local crime levels than national ones. Those that thought local crime had increased over the past few years felt more than twice as likely they would be a victim of crime in the next 12 months (33%), compared with 13% who did not. With regards to perceptions of crime in the country as a whole, those who perceived a rise over the past few years, were more likely to believe they would be a victim of crime than those that did not (23% and 13% respectively) (Table 2.6).

The number of those who perceived it likely they would be victimised in the next year varied across different groups within the population in similar ways to those outlined for perception of and worry about crime ( Appendix Table 2.04 and 2.05 (733.5 Kb Excel sheet) ). However, unlike worry about crime, there were no significant differences between the sexes in relation to perceived likelihood of victimisation.

These are some of the more notable variations:

• amongst those that had been a victim of crime in the last 12 months, 36% felt they were likely to be a victim again in the coming year compared with only 16% non victims

• those aged 65 and over, were less likely than all other age groups to expect to be a victim of crime in the next 12 months. It was estimated that, 15% of 65 to 74 year olds, and 10% of those aged 75 and above thought it “likely” they would be a victim of crime in the next year. This compared to 24% of 35 to 44 year olds

• those resident in an urban area were more likely to think they will be a victim of crime in the next year (21%), than those resident in a rural area (14%)

• residents in areas with high levels of physical disorder; were more likely to think they will be a victim of crime in the next 12 months (28%), than those living in areas with low physical disorder (19%)

• residents in the 20% most deprived areas were more likely (26%) than those resident in the 20% least deprived areas (15%), and those resident in all other areas (19%) to believe they will be a victim of crime in the next year.